Optimalisasi Metode Naive Bayes untuk Menentukan Program Studi bagi Calon Mahasiswa Baru dengan Pendekatan Unsupervised Discretization

Abstract

The admission of prospective new students must consider various procedures to direct prospective new students in determining the study program they are interested in. This study will discuss the optimization of the Naive Bayes method to determine the study program or major for prospective new students with the Unsupervised Discritization method approach. There are several stages of research methods carried out in this study, including Data Cleaning, Data Collection, Criteria Determination, Probability Determination, and Data Testing. This research has been carried out using the same method, namely the Naïve Bayes method which is used to classify the interests of prospective new students in determining the study program with an accuracy value of 96.68%. Ongoing research uses the same method, namely Naive Bayes, then optimization is carried out with the Unsupervised Discretization method approach. For data testing, there are 1671 student data records. After testing with the same method and optimizing it, the accuracy value from 96.68% became 97.66% with the classification results showing the DIII Pharmacy study program. The purpose of this research is to produce a classification in determining the study program or major for prospective new students using the Naïve Bayes method by the optimization of the Unsupervised Discretization method. From the results of testing the data, the Naïve Bayes method after optimization with the Unsupervised Discretization method is very good compared to the application before optimization.

Keywords

• interests
• naïve bayes
• unsupervised discretization
• study